Hey REW crew, had a question about RTA measurements.
This arose out of idle curiosity about measuring certain "low volume" Disney/Marvel movies, I wanted to objectively measure the response of certain movie scenes to get a sense of avg vs. peak levels. I've never really used the RTA feature and figured that was the best method to analyze the levels of actual program content over a period of time.
So I placed my UMIK-1 mic in the MLP, set my master volume to -10, and played certain test scenes with explosions and loud noises, and started the RTA. My system is pretty accurately calibrated with MV 0.0 set to reference level. The graphs below are my peak vs. avg RTA results for two scenes: (1) Thanos destroying the Avengers facility in Endgame, and (2) the first few minutes of Chapter 3 of Mad Max: Fury Road.
Not knowing much about the RTA settings, I left everything pretty much at defaults in REW. For the "peak" measurements below, I left averages at "none". For the averages I set averages to "Forever" assuming that would continuously average from the point at which I clicked "Reset Averaging" to the point at which I stopped "recording" with the RTA. There is no smoothing applied.
I found the results a bit surprising. The peak levels in the 95-100dB range (around ~30Hz) correspond fairly well to the "max" levels when I use the SPL meter (C-weighting), although the SPL max is a few dB higher. Both graphs show a fairly consistent ~10dB gap between the "average" and "peak" levels.
But what surprised me was how little energy the RTA shows above the deep bass region. Taking the Endgame example, it's peaking at ~75dB at ~100Hz, down to ~65dB at ~200Hz... at 1kHz it's averaging ~35-40dB and peaking at 50-55dB, and lower and lower from there. I had relatively consistent results with several different movies and scenes so it's not just a fluke of a particular piece of content I used.
Surely this can't be right, right?
Is there really such little energy up high? If dynamic content in the tweeter range (2kHz+) is below 50dB at -10 MV (which is pretty darn loud) then how does anyone ever blow a tweeter?
I feel like there's something I'm missing.
I can only assume that either (1) something is wrong with my technique (e.g. bad settings in REW) or (2) I am interpreting the results wrong or perhaps (3) I am not understanding what the RTA is supposed to be showing and I should be using a different tool.
Or maybe, this is accurate?? I know people have done spectrum / amplitude analysis of movie soundtracks using other tools (like in the BEQ threads)... anyone who understands the REW RTA feature have any feedback?
Avengers: Endgame (Thanos destroys the Avengers facility) at -10 volume:
Mad Max: Fury Road (first few minutes of Chapter 3) at -10 volume: